Drunk Monk
rrProfessor
Location: Preston, UK
My Posts This: Topic Forum | I only quoted that bit because that was the bit I thought you were talking about.
I think part of the problem is that people aren't fully aware of what the difference are between anolog, digital and super servo's. The best post I've ever read on the issue was by dkshema in November last year and it took some finding again. This will explain what a digital servo is and does a much better job of explaining it that I could ever do 
___________________________________________________
Digital servos will work with any standard RC receiver, PCM, PPM/FM, or otherwise. The servo does not know where it is being driven from, or by what kind of receiver. In turn, the receiver doesn't know what kind of servo is hooked up to it.
Your transmitter encodes stick position for each channel in the form of a short electrical pulse. Typically, the pulse width at neutral is about 1.5 milliseconds, and will vary with stick position from a minimum of 1 millisecond (say, full left cyclic), to a maximum of 2 milliseconds (full right cyclic). Each channel has its own pulse of varying width, determined by the stick or switch position. The transmitter assembles each of these pulses into a repeating "train" -- or chain. The pulses are always assembled in the same sequence (aileron always first, elevator next, then throttle, rudder and the rest of the channels -- the order may vary from manufacturer to manufacturer). An entire set of pulses for all channels is called a "frame". The spacing between pulses in each frame is fixed, the width of each pulse varies with the stick position. Frames are separated from each other by a longer, fixed amount of time. The rate at which each frame is transmitted is called the frame rate. The frames have historically been transmitted at somewhere between 50 and 60 frames a second. This "frame rate" determines how often a servo gets a new position update. Each frame represents a new position update, therefore, each servo will get new information at about 50 to 60 times per second.
The receiver's job is to decode the pulse width information in the transmitted signal, and send the proper pulse to the proper servo.
Looking at a particular channel, left/right cyclic for example, the servo connected to that channel will try to position its output arm to match the width of the pulse it is receiving from the transmitter via the receiver.
Inside an analog servo, there is a circuit that is constantly generating a 1.5 millisecond wide "reference" pulse (remember from above, 1.5 milliseconds is considered to be the neutral position). The width of the incoming pulse from the receiver is compared to that of the reference pulse. If the two pulse widths are the same, the servo does not move. If the incoming pulse width is narrower than the reference pulse (say 1 millisecond), the servo electronics applies power to the servo motor which starts the motor turning. Through a series of gears, you get the motion of the arm or wheel that you desire, but the gear train is also connected to a variable resistor, or potentiometer ("pot" for short). As the motor and gears turn, the resistance of the pot changes. The variable resistance value is used to adjust the width of the "reference" pulse in the servo. As the motor and gears turn, the width of the reference pulse is adjusted until the two pulse widths match, at which time the motor stops, and you have a new output wheel position. The servo gets a new position update at the transmitter frame rate -- 50 to 60 times a second. Between updates, with no new position information, the motor is not powered. If you try to move the output arm (or flight loads are present) this will affect the setting of the pot and the electronics will apply power to the motor to keep it at it's current position.
The analog electronics smooths out the pulses of power being applied to the motor using a circuit called a pulse stretcher. The electronics also has some additional feedback built in to keep the servo from "overshooting" its intended position. This part of the feedback loop provides what is called "deadband" -- a small window of time where the motor is not driven. If the feeback loop is too "tight", the servo will continue to "hunt" around its current position. If the feedback loop is too "loose", the servo will overshoot its intended position, and the output arm will tend to "bounce" back and forth around its final position for a short period. The deadband is tweaked by the designer to provide a servo that is fast, doesn't "overshoot" or "undershoot" its position when moving, and so that it doesn't continually oscillate (buzz) when at rest. The deadband introduces a small amount of slop in the system -- you must move the control stick on the TX a certain amount to overcome the dead zone.
In a digital servo, the analog processing stuff has been replaced with a microprocessor. This little computer monitors the incoming pulse width -- it samples (digitizes) the pulse from the RX.
There is still a reference pulse being generated inside the servo, hence, still a need for the feedback pot. The reference pulse width is also digitized by the processor in the servo.
The frame rate from the transmitter is still 50 to 60 times per second, the incoming pulse width to the servo is still in the 1 to 2 millisecond range. The servo is still seeing a new position only 50 to 60 times per second from the transmitter/receiver.
As with the analog servo, the idea is to match the incoming pulse width from the receiver to that of the reference pulse in the servo. The microcomputer controls the switching of power to the motor. When the two pulse widths differ, the computer applies voltage to the motor, monitors the value of the feedback pot, and then stops the motor when the two pulses again match.
The digital servo has the distinct advantage of "knowing" how far the motor must move to get to its new position (the processor knows current position, desired position, and can calculate what it takes to move quickly and efficiently). The processor can drive the motor rapidly for large position changes, and accurately know when to slow down to prevent over or undershoot of the new position. Deadband is still present, but it can be much more accurately controlled by the computer. The control loop inside the servo can be much more responsive and accurate using digital techniques over the analog processing.
The digital servo also has the advantage that the processor can continue to provide shots of current to the motor between position updates (something not easily done in an analog servo) to hold the motor/gear train in its current position until a new position update is received (this is why digital servos will "buzz" when the sticks aren't being moved. The motor is being held in position, instead of "hunting" around its final position.
Analog servos would buzz if the deadband feedback is too tight. This was bad, because it meant the motor and feeback pot were continually moving a small amount -- this led to the feedback pot element wearing, getting crud on it, and eventually making the servo useless until it was disassembled, the pot cleaned and lubed, and put back together.
Digital servos have the distinct advantage of being able to achieve high transit speeds due to the "smarts" designed into the processor program, and high holding torque (since the motor is driven continually between position updates). Accuracy and repeatability are also improved over analog servos.
The downside of digital servos? Well, since the processor is now providing those extra pulses of power to the motor to improve the holding torque and position accuracy between pulse updates, you're using more current out of the battery to run the servo. Your batteries just don't last as long between charges.
The extra current required to achieve low transit times requires a larger gauge wire in the servo lead to reduce losses in the wire due to its resistance. You'll need to make sure servo extensions (if you need them) are also made with the larger gauge wire). Battery packs should also be equipped with larger gauge wires to minimize voltage drop in the leads when all those servos are moving as you'e doing your 3-D stuff with wild abandon.
This results in your needing to either buy higher capacity batteries for your flight pack, limit your flight times and number of flights each day, or go out and purchase a fast field charger, and battery monitor so you can pump up your flight pack while you're at the flying site.
Sorry this is so long winded, but then you did ask what the difference was. Now you know.
Stephen
I only open my mouth to change feet..... |
As far as I'm aware my Raptor 30 is in PPM.
